# US 2026/0305578 A1 · Multi-channel walk-in floor-reared poultry growth performance measurement system and method

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2026/0305578 A1
- Document ID: US20260305578A1
- Status: Published application
- Kind: A1
- Application number: 19/296,115
- Filed: 2025-08-11
- Published: 2026-10-08
- Earliest claimed: 2025-04-08
- Assignees: South China Agricultural University; Guangdong Laboratory for Lingnan Modern Agriculture Science and Technology; Guangzhou Huanong University Smart Agriculture Technology Co., Ltd.
- Applicants: South China Agricultural University; Guangdong Laboratory for Lingnan Modern Agriculture Science and Technology; Guangzhou Huanong University Smart Agriculture Technology Co., Ltd.
- Inventors: Deqin Xiao; Shaohai Peng; Dakang Guo; Zhuoheng Tang; Lei Cheng; Zeyuan Lin; Weichao Lin; Yuanyang Mao; Jiatao Wang; Handong Wang; Weiming Liu; Ruofei Tan
- Web page: https://patentgrep.com/patent/US20260305578A1

## CPC

- A01K 29/002: {for monitoring or measuring size or weight of the animal, e.g. monitoring growth}
- A01K 29/005: {Monitoring or measuring activity}
- A01K 31/18: Chicken coops or houses {for baby chicks}; Brooders {including auxiliary features, e.g. feeding, watering, demanuring, heating, ventilation (battery breeding cages A01K31/005)}
- A01K 35/00: Marking poultry or other birds {(marking meat A22C17/10)}
- G01G 3/142: Circuits specially adapted therefor
- G01G 17/08: for weighing livestock
- G01S 1/68: Marker, boundary, call-sign, or like beacons transmitting signals not carrying directional information
- A01K: ANIMAL HUSBANDRY; AVICULTURE; APICULTURE; PISCICULTURE; FISHING; REARING OR BREEDING ANIMALS, NOT OTHERWISE PROVIDED FOR; NEW BREEDS OF ANIMALS
- G01G: WEIGHING (sorting by weighing B07C5/16)
- G01S: RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES

## Abstract

A multi-channel walk-in floor-reared poultry growth performance measurement method utilizes a walk-in, multi-channel growth performance measurement device based on behavioral characteristics. This design adapts to different scales of poultry farming, allowing for flexible expansion and deployment. Through a signal calibration matching strategy, the RFID signals from various channels are accurately matched with strain gauge sensor data; this ensures that signals between different channels do not interfere, enabling error-free individual information identification. By using RFID technology to record poultry's activity frequency and dwell time in different areas, we can analyze their activity distribution patterns; this provides data support for further behavioral research and health status monitoring, offering a large amount of effective data for screening superior poultry; furthermore, this system achieves unmanned automated monitoring for the farm, improving poultry farm production efficiency and reducing flock stress caused by manual operations, guaranteeing the standardization and reliability of data collection.

## Claims

### Claim 1

A multi-channel walk-in floor-reared poultry growth performance measurement method, comprising the following steps:
S1: constructing a multi-channel walk-in floor-reared poultry growth performance measurement system, comprising a control processing module, a position information module, a behavior analysis module, a weight calculation module, and a result storage module,
the position information module unidirectionally transmits poultry position information to the control processing module, the control processing module processes data to obtain position information of the poultry; the control processing module then sends an instruction to activate the weight calculation module, the weight calculation module collects dynamic weight of the poultry passing through the channel and calculates true weight of the poultry before returning it to the control processing module; the control processing module obtains behavior type of the poultry from the collected information via the behavior analysis module, finally, based on the obtained dynamic weight and behavior characteristics of the poultry, predicted weight of the poultry is analyzed, and the data is transmitted to the result storage module for saving;
building a multi-channel walk-in floor-reared poultry growth performance measurement device, comprising a ramp, an RFID reader/writer, a channel platform, a bottom resistive strain gauge sensor, channel railings, an RFID reader/writer, a system host, a multi-function weighing and force measuring instrument, a system display screen, a console-type operating enclosure platform, and channel baffles;
S2: when the poultry steps onto the channel platform, the weight calculation module activates, at this point, the resistive strain gauge sensor at a bottom of the channel platform begins to operate, continuously reading weight information of the poultry, the weight reading process is completed when the weight returns below a threshold;
S3: calculating the true weight of the poultry from the collected weight information, the specific method is as follows: S31: calculating expected value of overall dynamic weight data from the continuously read poultry weight information obtained in S2;
S32: determining if the dynamic weight is stable based on the expected value calculated in S31, if not stable, filtering out abnormally fluctuating body weight values by using a median filtering algorithm;
S33: performing mean filtering on median-filtered weight values to reduce overall dynamic weight fluctuations;
S34: based on processing results from S33, inputting processed data into the weight calculation module for weight calculation and outputting weight calculation results;
S35: dividing the weight calculation results output from S34 into multiple data slices, and calculating slope, intercept, and root mean square error for each group of data;
S36: repeating S31-S35 until there are no further data groups, then the weight calculation is completed, and an optimal predicted weight is output as the true weight of the poultry;
S4: calculating distribution locations of the poultry based on read identity information of the poultry.

### Claim 2 (depends on 1)

The multi-channel walk-in floor-reared poultry growth performance measurement method of claim 1, wherein a calculation method for the expected value in S32 is as follows:
E
=
1
n
-
1
⁢
∑
i
=
1
n
-
1
❘
"\[LeftBracketingBar]"
w
i
+
1
-
w
i
❘
"\[RightBracketingBar]"
wherein E represents the expected value of dynamic weight curve, n represents the number of weight data collected, wi+1 and wi respectively represent true weight values at time i+1 and time i.

### Claim 3 (depends on 1)

The multi-channel walk-in floor-reared poultry growth performance measurement method of claim 1, wherein a specific method for the median filtering in S32 is as follows:
?
=
median
(
w
i
-
m
,
w
i
-
m
+
1
,
…
,
w
i
+
m
)
wherein i represents time, represents predicted weight value, wi represents true weight value, m represents a window size of the median filtering, median ( ) represents a calculation of median for a set of data.

### Claim 4 (depends on 1)

The multi-channel walk-in floor-reared poultry growth performance measurement method of claim 1, wherein a specific method for the mean filtering in S33 is as follows:
w
^
i
=
1
2
⁢
m
+
1
⁢
∑
k
=
i
-
m
i
+
m
w
k
wherein i represents time, represents predicted weight value, k represents a position of weight data within the filtering window, wk represents weight value before mean filtering, m is a window size of the mean filtering.

### Claim 5 (depends on 1)

The multi-channel walk-in floor-reared poultry growth performance measurement method of claim 1, wherein a specific method for the root mean square error in S35 is as follows:
RMSE
=
1
M
⁢
∑
i
=
1
M
(
w
^
i
-
w
i
)
2
wherein i represents time, represents predicted weight value, wi represents true weight value, M represents the number of samples used to calculate the root mean square error.

### Claim 6 (depends on 1)

The multi-channel walk-in floor-reared poultry growth performance measurement method of claim 1, wherein S4 specifically comprises the following steps:
S41: after the poultry passes through the channel, information collected by the RFID reader/writer is transmitted to the position information module, the information is then sorted by the order of collection time;
S42: after preprocessing the data, RFID antenna numbers from which signals were read before and after the poultry passed through the channel are compared to determine if the poultry has indeed gone through that channel;
S43: based on the RFID antenna numbers where identity information was read (before and after), direction of entry/exit of the poultry is determined; according to the currently read RFID antenna position of the poultry and the entry/exit direction, current location information of the poultry is determined;
S44: by comparing relevant data (collected time, weight, and reading position) during different time periods, a distribution pattern of the poultry is obtained.
